丹麦版BACQ-14扩展版心理测量学特性研究：从12项到14项的四维应对量表

Abstract

Objective:

To describe the expansion of the three-dimensional Brief Approach/Avoidance Coping Questionnaire (BACQ) from a 12-item scale (BACQ-12) to a 14-item scale (BACQ-14) covering four dimensions and to examine its clinimetric properties including reliability and validity.

Methods:

A thorough developmental process for the BACQ-14 was performed. Pilot tests were conducted to evaluate content validity. A field test comprising both test and retest was performed. Structural validity was evaluated using Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA). Internal consistency was evaluated by Cronbach’s alpha and item-to-rest Pearson correlation. Hypothesis testing was based on health-related variables, whereas test–retest reliability was evaluated using Bland–Altman (BA) plots.

Results:

The BACQ-14 showed good content validity. The EFA (n = 150) showed that both a three- and a four-factor-model could be suitable for the BACQ-14, but the CFA (n = 360) demonstrated that the four-factor model had a substantial better fit. Cronbach’s alpha for the dimensions in the four-factor model ranged from 0.66 to 0.88. The BA plots showed that the 95% limits of agreements were approximately [−5;5] for the four subscales, indicating large random variation.

Conclusion:

The BACQ-14 covering four dimensions had slightly better clinimetric properties than the BACQ-12 and is suitable for questionnaires where a brief assessment of coping is warranted. However, its sensitivity to detect small changes in coping is constrained which may limit its utility in clinical practice.

Introduction

Many definitions of coping exist, but according to Lazarus and Folkman coping covers the cognitive and behavioural efforts made by an individual to manage external or internal demands associated with a stressor that seems to tax or exceeds the individual’s resources (Lazarus and Folkman, 1984). This definition emphasises that coping includes both cognitive processes and behavioural efforts in response to stress.

More than 400 different coping strategies have been identified (Skinner et al., 2003). They are often classified into overall distinctions such as problem- versus emotion-focused, cognitive versus behavioural, and approach versus avoidance (Skinner et al., 2003; Carver and Connor-Smith, 2010; Jacobs and Carver, 2020). These distinctions are not exclusive, as the strategies depend on the purpose. For instance, seeking support can be regarded as both emotion-focused and problem-focused coping, depending on whether the purpose is to obtain emotional support or seek advice and instrumental help from others (Carver and Connor-Smith, 2010).

Coping affects health outcomes (Penley et al., 2002; Moskowitz et al., 2009; Suls and Fletcher, 1985; Rutter and Rutter, 2002), thus valid tools for measuring coping are valuable. Several scales for measuring coping are available (Krohne et al., 2000; Endler and Parker, 1994; Amirkhan, 1990; Folkman and Lazarus, 1988), however, most of them include at least 30 items, which makes them difficult to implement in both clinical practice and research studies. The original Brief Approach/Avoidance Coping Questionnaire (BACQ) developed by Finset et al. (2002) includes 12 items and has been used in several surveys covering multiple constructs (Soberg et al., 2007; Soberg et al., 2015; Loton et al., 2016; Rutskij et al., 2010; Allen, 2021; Ulrichsen et al., 2021). The original BACQ measures coping in the approach-avoidance dichotomy, where Avoidance can be divided into Resignation and Withdrawal and Diversion (Finset et al., 2002). The BACQ was translated and adapted into Danish as part of a large population-based survey called the Danish Symptom Cohort (DaSC) in 2012, which was repeated in 2022 (Rasmussen et al., 2014; Sætre et al., 2024). Examination of the clinimetric properties of the Danish BACQ showed that a three-factor model was acceptable, but that a four-factor-model may perform even better (Sætre et al., 2023). Yet, the fourth factor suggested in the factor analyses only consisted of two items with high ceiling effects, poor factor loadings, and low item-to-item correlations (Sætre et al., 2023). Therefore, adjustments of the scale were necessary to enable the use of the Danish BACQ as a four-factor model. Thus, we sought to develop and improve the performance of the fourth dimension by revising and expanding the original 12-item BACQ (BACQ-12). This resulted in a 14-item questionnaire named the BACQ-14. Further, the introductory text was considered difficult to understand by several participants in the 2012 DaSC study. Hence, efforts were made to improve it as part of the development of the BACQ-14. The aims of this study were (1) to describe the expansion of the BACQ-12 to the BACQ-14 and (2) to examine the clinimetric properties of the BACQ-14 including reliability and validity.

Materials and methods

The BACQ-14 was developed based on the BACQ-12, which is a reflective, multidimensional scale consisting of 12 items within three subscales: Approach (items 1–6), Diversion (items 7–9), and Resignation (items 10–12). The development included three steps: (1) development of a fourth dimension, (2) improvement of the introductory text, and (3) examination of the clinimetric properties of BACQ-14.

Development of the fourth dimension

The development of the fourth dimension was subdivided into another three steps: (1) Definition and elaboration of the construct to be measured, (2) a literature search, and (3) selecting and formulating items.

Definition and elaboration of the construct

The inadequate fourth dimension found when examining the clinimetric properties of the BACQ-12 comprised the two items: (1) “I say so if I am angry or sad” and (2) “I like to talk with a few chosen people when things get too much for me” (Sætre et al., 2023). For the purpose of defining and elaborating the construct of the new dimension, we established an expert panel consisting of six persons including three medical doctors who were all part of the project group of the Danish Symptom Cohort II (DaSC II) in 2022, a psychologist with profound knowledge about coping, a person with a master’s in Religion with expertise in communication and relationships, and an anthropologist who had conducted interviews during the development process of the DaSC II questionnaire. From these interviews, she had obtained knowledge about the interpretation of and immediate responses to the BACQ-12. Four members of the expert panel had comprehensive knowledge about the Danish BACQ-12 from the validation procedure (Sætre et al., 2025). The panel also had extensive knowledge of questionnaire development and clinimetrics in general from previous work.

Literature search

According to Finset et al. (2002), the two items were originally placed in the socioemotional domain of the BACQ-12. Based on this knowledge and considering the wording of these two items, the expert panel decided that the construct of the fourth dimension should cover issues related to communication, verbalisation, and expression of feelings.

A literature search was conducted to find relevant questionnaires covering the construct, as no scales should be invented if other scales covering the same construct existed (Terwee et al., 2011). No sufficient instruments were revealed in the literature search, yet a few instruments were saved for later inspiration (Whitman, 1929; Carver, 1997).

Selecting and formulating items

An item bank was established based on reflections from the expert panel, discussions within the project group, and the existing literature. The meaning and wording of each item were reviewed carefully by the project group.

Improvement of the introduction text

Although the DaSC study was not directly a part of the development of the BACQ-14, the comments to the DaSC questionnaire written in the free text box in the final section of the questionnaire provided valuable information about the interpretation of the BACQ-12 (Rasmussen et al., 2014; Sætre et al., 2023). For example, several respondents were in doubt about which situation they were supposed to think of when answering the questions and stated that they tended to use different strategies in different situations. Also, some respondents stated that they never experienced any problems. Consequently, the introductory text of the scale was reviewed concomitantly with the development of the fourth domain.

Examination of the clinimetric properties of BACQ-14

We followed the COnsensus-based Standards for the selection of health Measurement INstruments (COSMIN) taxonomy for evaluating the clinimetric properties of the BACQ-14 taxonomy (Mokkink et al., 2010b; Mokkink et al., 2010). Pilot tests were performed to examine content validity including face validity, and a field test including a retest was carried out to examine construct validity, including structural validity, and hypothesis testing as well as reliability and stability.

Pilot tests

Two rounds of pilot testing were performed. First, a preliminary version of the BACQ-14 was distributed by email to individuals within our academic environments, family, and friends. Respondents were asked to comment on the introductory text and the first four items. The written comments were collected and discussed within the project group. Second, the face validity of the BACQ-14 was examined in a qualitative pilot test in 11 informants from the general population: four men and seven women aged 21–90 years. Each informant was given a printed version of the questionnaire and asked to think aloud while completing it. Simultaneously, an interviewer observed and noted non-verbal reactions. Afterwards, the informants were interviewed and asked to elaborate on their thoughts and feelings during the questionnaire. Further, they answered some predefined questions including whether the introduction text was comprehensible, and whether the informant thought about a single specific situation throughout the questionnaire or different situations.

Field test

We conducted the field test by distributing the survey amongst individuals 20 years or older in a test–retest set-up. The study population was recruited through a research agency [Forskningnu.dk] that distributed a universal link on social media and amongst individuals from their database. The database comprised previous participants in research projects including both healthy and non-healthy individuals of different ages. Although it was not possible to obtain a representative sample of the Danish general population, an effort was made to ensure a balanced distribution of men and women and to recruit participants across different ages. The link provided access to a questionnaire on the secure survey platform SurveyXact. The questionnaire was initiated by a description of the study, including the participants’ legal rights, and had a forced answer structure to avoid missing answers.

To enable evaluation of the stability of the BACQ-14, participants were asked permission to receive an invitation for the retest questionnaire. To encourage participation, we drew lots for three gift boxes containing two movie tickets amongst those who completed the retest questionnaire. Participants who consented to participate twice received an invitation in their personal email inbox two-three weeks after completion of the first survey. The invitation to the retest was distributed in three waves to ensure proper time lap between the two surveys. Data collection was conducted from May to June 2023.

Test–retest-reliability of the BACQ-12

Test–retest-reliability of the BACQ-12 has never been performed, neither as part of the development of the original BACQ-12 (Finset et al., 2002) nor as part of the validation of the Danish version (Sætre et al., 2025). Since we had already established a test–retest setup, we decided to include the BACQ-12 in both the initial questionnaire and the retest questionnaire.

Covariates

In addition to the BACQ-14 and the BACQ-12, questions about various sociodemographic characteristics were asked to obtain information about sex, age, educational level, attachment to labour market, and ethnicity. Additionally, the respondents were asked to indicate whether they had a chronic disease and to provide information about their level of concern for their health, self-rated health, and physical fitness on five-point Likert scales. The exact wording of the questions can be found online (Syddansk Universitet, 2024).

Statistical analyses

Study sample and subsamples

Only individuals who had completed the initial questionnaire were included in the study sample. Respondents who had ticked the same answer to all questions of either the BACQ-12 or BACQ-14 were excluded. This was done based on the assumption that some variation in replies must be expected across subscales (Skinner et al., 2003; Folkman et al., 1986; Connor-Smith and Flachsbart, 2007). For the purpose of analysis, the study sample was divided into three subsamples: Sample A, Sample B, and the Retest sample. Those who solely completed the initial questionnaire were split randomly into Sample A and B, respectively, while the Retest sample consisted of all participants who completed the retest questionnaire.

Scoring of the BACQ-14

Similar to the BACQ-12, all items of the BACQ-14 were scored on a 5-point Likert scale ranging from 1 (Disagree completely) to 5 (Agree completely). Mean score, standard deviation (SD), and response rates for each response category were calculated for all items.

Construct validity

The construct validity of the BACQ-14 was examined by means of structural validity and hypothesis testing [27, 28]. Structural validity included Exploratory Factor Analysis (EFA) based on Sample A, and Confirmatory Factor Analysis (CFA) based on Sample B and the Retest sample.

Exploratory factor analyses (sample a)

Prior to the EFA, a Kaiser-Meyer-Olkin (KMO) test of sampling accuracy was performed to assess whether the sampled data were adequate for such analyses. KMO measures were interpreted using the cut-offs suggested by Kaiser and Rice (1974): Unacceptable: 0.0–0.49, Miserable: 0.50–0.59, Mediocre: 0.60–0.69, Middling: 0.7–0.79, Meritorious: 0.80–0.89, and Marvellous: 0.9–1.00. Afterwards, the EFA was conducted on Sample A. We assessed the suitable number of factors in the BACQ-14 by generating a scree and parallel plot (Dinno, 2010). Factors with eigenvalues above the parallel line were included in the EFA (Ledesma et al., 2015).

Confirmatory factor analyses (sample B)

The CFA was conducted on Sample B and the Retest sample to confirm the models suggested in the EFA using Stata’s framework of Structural Equation Models (SEM) for linear models (StataCorp, 2019). As a sensitivity analysis, we performed both EFA and CFA on polychoric correlations. The precise syntax is included in the Supplementary Table 2. For comparison (regardless of the EFA results) we decided to include the three-factor solution suggested for the BACQ-12. The following goodness-of-fit indices were used: Root Mean Squared Error of Approximation (RMSEA, Satorra-Bentler scaled), Tucker-Lewis Index (TLI, Satorra-Bentler scaled), Comparative Fit Index (CFI, Satorra-Bentler scaled), Standardised Root Mean Squared Residual (SRMR), Akaike’s Information Criteria (AIC), and Bayesian Information Criterion (BIC). All reported estimates are standardised. The following cut points were applied to indicate a good fit: RMSEA <0.08, CFI ≥ 0.90, TLI ≥ 0.95, and SRMR < 0.08 (Hooper et al., 2007; Kline, 2005).

Reliability

Reliability of the BACQ-14 was examined in terms of internal consistency (Cronbach’s alpha and Mc Donald’s Omega), reliability (item-to-rest Pearson correlation), and test–retest reliability (Bland–Altman (BA) plots). The BA plots illustrate the absolute differences between score measurement with 95% limits of agreements (Qin et al., 2019; Altman and Bland, 1983), and these were prepared separately for all four subscales of the BACQ-14 and all three subscales of the BACQ-12. To evaluate test–retest reliability further, the BA plots were supported by Pearson correlation coefficient, intraclass correlation coefficient (ICC) based on a linear full maximum-likelihood random-intercept model (absolute agreement), concordance correlation coefficient Rho (Lin, 1989), average difference between measurements (Diff: average) and the standard deviation of the difference (Diff: SD), standard error of measurement of agreement (SEM-agreement) as Diff: SD divided by the square root of 2, and the coefficient of repeatability (CR). The CR is defined as the half-width of the normal-based 95% prediction interval for the within-subject difference, calculated by 1.96 x square root of 2 x SEM-agreement (or standard deviation of the difference) and equals the half-length of the Bland–Altman limits of agreement (LOA). With no expected mean difference between two measurements, observations are expected to vary randomly around zero up to +/− CR.

We used Stata version 18 or later, and p-values <0.05 were considered to be statistically significant. Even though we state hypotheses, we regard our analyses and corresponding p-values as descriptive rather than confirmatory. Hence, we employed no correction for multiple testing.

Hypotheses testing by comparison of subgroups

For hypothesis testing, sum scores were calculated for each of the four dimensions of the BACQ-14; items n1-n4 (the new dimension) called Verbalisation, n5-n8 (i.e., the remaining approach items of BACQ-12) termed Action, n9-n11 termed Diversion, and n12-n14 termed Resignation. The rationale for using Verbalisation and Action is described in the results section. For Verbalisation and Action, the total sum score ranged 4–20 and for Diversion and Resignation, the total sum score ranged 3–15.

Based on the literature and clinical experience, we a priori hypothesised that individuals with good self-rated health, low concerns about their current health, and good physical fitness would have the highest scores on Verbalisation and Action and the lowest scores on Diversion and Resignation (Penley et al., 2002; Suls and Fletcher, 1985; Folkman et al., 1986; Frølund Pedersen et al., 2016; Raasthøj et al., 2022; Panayiotou et al., 2014). The hypotheses were assessed by calculating mean sum scores for each dimension in the subgroups within each covariate, and differences between the subgroups were examined by linear regression models. For comparison of the subscales with different score ranges, we calculated means and corresponding 95% CI for transformed scores, where all scores were transformed to a uniform scale ranging from 0 to 100. Additionally, we calculated means and corresponding 95% CI for z-scores, offering the possibility to compare effect sizes (group means) in terms of SD.

Covariates included in the analyses were sex (women/other and men); age groups (20–39 years, 40–59 years and 60–79 years); self-rated health (poor, fair, good, very good, and excellent); concern about current health (not at all, slightly, moderately, quite a bit/extremely); physical fitness (poor, not so good, fair, good, very good); attachment to labour market (employed, unemployed, post-employment/retirement), chronic disease (yes, no/do not know), educational level (No/other, Low (< 10 years), Middle (10–15 years), High (> 15 years)); ethnicity (Danish, Other than Danish/prefer not to answer). These variables were selected based on their influence on coping (Lazarus and Folkman, 1984; Kirmayer and Sartorius, 2007; Kirmayer, 2005; Taylor and Stanton, 2007).

Results for respondents who selected other for sex were not reported separately because groups with fewer than five individuals cannot be presented due to confidentiality requirements in accordance with Danish Data Legislation. Consequently, these respondents were combined with the largest category, women.

Ethics statement

The introduction of the questionnaire included detailed information about the objective of the study and legal rights. The participants were informed that participation in the study was voluntary, and in case of any inquiries or clarifications about the study, participants could contact the study group via email. The study was conducted as part of the DaSC which has been approved by the Research Ethics Committee at University of Southern Denmark (Case no. 21/29156) and registered by the Danish Data Protection Agency (j.no. 2011–41-6651) through the Research and Innovation Organisation (RIO), University of Southern Denmark (Project number 10.104).

Results

Conceptualisation of the construct in the fourth dimension

During the process, the expert panel decided on the concept of the new dimension concerning communication, verbalisation, and expression of feelings. This led to a thorough discussion of different factors influencing these aspects such as personality (extroversion vs. introversion), the form of communication (oral vs. written), resources (e.g., social network), the number of people you speak with (many vs. few), and the confidentiality of the person you share with (acquaintance vs. close friends). Further, the expert panel discussed the purpose with expressing problems and feelings to other people. Some may involve others to seek support for finding a solution to the problem and/or to get emotional relief. Therefore, it was considered important that the new dimension covered both emotion-based and problem-based strategies as well as reflected approach coping in contrast to avoidance.

A member of the expert panel reflected that the BACQ-12 was horizontal in terms of communication, meaning that it only considered communication with other people, however, some people need a more vertical communication, e.g., in a spiritual or religious ways. Adding the vertical aspect was discussed within the project group, and a scale of religious and spiritual coping and conscious presence and self-control was reviewed for inspiration (Büssing et al., 2013). However, we concluded that the vertical communication could not be covered sufficiently by the current scale when complying with our intention to maintain briefness. To keep the possibility of vertical communication, the questions were phrased more openly such as “In difficult situations it helps me to share my feelings with someone”. In Danish, the pronoun “someone” [nogen] is not restricted to physical persons.

Based on the conceptualisation, the new dimension was termed Verbalisation, and the original dimension Approach was renamed Action because the four remaining items in the subscale were placed within the action-related and cognitive domains by Finset et al. (2002).

Selecting and formulating items

To maintain the briefness and symmetry of the scale, the new dimension should not contain more than four items. According to our conceptualisation, two items covering emotion-based coping strategies and two items covering problem-based coping strategies were suggested. The meaning and wording of each item within the item bank was discussed, and four items were finally selected and included in the test of the questionnaire.

Improvement of introduction text

Based on the definition by Lazarus and Folkman suggesting that coping is initiated when external or internal demands exceed the individuals’ resources (Lazarus and Folkman, 1984), the expert panel concluded that coping can deal with issues from small everyday banalities to larger life crises depending on the person and her/his resources. Therefore, giving a specific situation to think of could be an unintended limitation or introduce an unintended bias towards specific situations and unintentionally exclude others. We aimed to measure general coping and therefore, inspired by The brief COPE by Carver et al. (1989), we added that different strategies may be applied to different situations, but the respondents should think of how they usually tend to react.

Pilot test

The first pilot test resulted in feedback from 22 individuals. Their feedback was overall positive stating easy comprehensibility, yet some unclarities and rewordings of the four new items were suggested, which resulted in minor modifications. The second pilot test showed that the introductory text and the four new items were very well understood as most informants immediately completed the BACQ-14 without any problems. A few informants asked if they were supposed to think of a specific situation throughout the questionnaire, but when elaborating on their perception, they assumed that they should report how they generally coped. The follow-up interviews confirmed that all the informants thought of general coping rather than a specific situation. The second pilot test further indicated that items n7 and n8 prompted reflection and some uncertainty amongst participants, as responses were perceived to depend heavily on the specific situation. The items were “I think something positive could come out of my complaints/problems” (n7) and “I firmly believe that my problems will decrease (and my situation improves)” (n8). Although the project group initially did not intend to revise items outside the newly developed dimension, it was agreed that item 8 could be improved by removing the parenthetical phrase, thereby making it more straightforward and easier to interpret. No additional changes were made to items within the other factors.

The final wording of the BACQ-14, translated from Danish into English following Spine guidelines (Beaton et al., 1976), is shown in Table 1.

Table 1

Wording of the final BACQ-14.

*The parenthesis in old item 6 (new item 8) has been removed.

Field test

In total, 630 individuals initiated the first test questionnaire, however, 120 were excluded, mostly due to incomplete answers (Figure 1). In total, nine individuals had participated twice (revealed by their e-mail address being reported for retest), and another seven individuals had completely similar single-item answers throughout the BACQ-12 or the BACQ-14. This resulted in 510 eligible respondents in the study sample, which was divided into three subsamples: Sample A (n = 150), Sample B (n = 151), and the Retest sample (n = 209). Descriptive statistics of each sample is shown in Table 2. All subsamples had respondents of different ethnicity, educational level, and employment status. The three subsamples did not differ significantly except for the proportion of respondents aged 60–79 years, which was higher in the Retest sample (Table 2).

Figure 1

Table 2

Characteristics of the total test sample, including Sample A, Sample B, and the retest sample. The p-values are based on Chi-squared tests.

¹Sample A consists of a random half of respondents completing only the test-questionnaire.

²Sample B consists of the other random half of respondents completing only the test questionnaire.

³The Retest sample consists of all respondents completing both the test and retest questionnaire.

⁴Less than 5 reported “Other” for sex. Therefore, these respondents are grouped with women, which is the largest group.

⁵The categories “No” and “Other” in education and “Do not know” in chronic disease were not included in the chi^2 test due to small numbers.

Item characteristics of the new dimension of BACQ-14

Table 3 shows the distribution of answers for each item of the BACQ-14. For the four items in Verbalisation, all response categories were represented, although there were ceiling effects with high response rates for Completely agree ranging from 17.6% for n3 to 35.7% for n1. The mean item scores ranged from 3.44 (SD 1.17) for n3 to 4.00 (SD 1.03) for n1.

Table 3

Distribution of answers per item and response category for the BACQ-14 (n = 510).

^*SD, standard deviation.

Construct validity

Exploratory factor analyses (Sample A)

The KMO showed that most items were considered Middling or Meritorious, while only three (n6-n8) were considered Mediocre, and therefore the sampled data were adequate for the EFA. In the scree plot, three factors had an eigenvalue above 1, while the parallel plot showed at least four factors above the elbow of the graph (Figure 2). Therefore, both the two-, three-, and four-factor models in the EFA are presented in Table 4. Factor loadings above 0.3 were considered acceptable (Hair et al., 1998).

Figure 2

Table 4

Results of the Kaiser-Meyer-Olkin (KMO) test and exploratory factor analyses on Sample A (n = 150).

Factor loadings >0.3 in bold.

^*Uniq, Uniqueness.

KMO: Cutoffs: Unacceptable: 0.0–0.49, Miserable: 0.50–0.59, Mediocre: 0.60–0.69, Middling: 0.7–0.79, Meritorious: 0.80–0.89, and Marvellous: 0.9–1.00.

In the two-factor model, factor loadings were <0.3 for n5-n8. The three-factor model showed that n1-n4 had factor loading ranging from 0.78 to 0.87 on factor I, n5-n8 had factor loadings ranging from 0.46 to 0.75 on factor II, while n9-n14 had factor loadings ranging from 0.51 to 0.78 on factor III. In the four-factor model, n1-n4 had factor loadings ranging from 0.78 to 0.87 on factor I, n5-n8 had factor loadings ranging from 0.47 to 0.74 on factor II, n9-n13 had factor loading ranging from 0.34 to 0.83 on factor III, and n12-n14 had factor loadings ranging from 0.51–0.59 on factor IV. Thereby, n12 and n13 loaded on both factors III and IV although with a higher loading on factor IV.

Confirmatory factor analyses (Sample B and retest sample)

The two-factor model was not examined further since n5-n8 did not significantly load on any of the two factors. Goodness of fit indices for both the three- and four-factor models are shown in Table 5, and the four-factor model illustrated in the SEM in Figure 3. Goodness-of-fit indices were slightly better for the four-factor model compared with the three-factor model with RMSEA = 0.058 (<0.08), CFI = 0.953 (≥0.90), and SRMR = 0.065 (<0.08) (Table 5). AIC and BIC were also slightly lower for the four-factor model compared with the three-factor model.

The results based on polychoric correlations are reported in Supplementary Table 1 (EFA) and Supplementary Tables 2 and 3 (CFA). Changing the framework of Structural Equation Models (SEM) for linear models to polychoric correlations overall showed similar results.

Table 5

Estimation results of confirmatory factor analysis on Sample B and retest sample (n = 360): Goodness of Fit Indices (Satorra-Bentler scaled).

Levels for good fit: Root Mean Square Error of Approximation (RMSEA): <0.08, Comparative Fit index (CFI): ≥0.90, Tucker Lewis Index (TLI): ≥0.95, Standard Root Mean Square Residual (SRMR): <0.08.

Figure 3

Internal consistency

Table 6 shows the item-to-rest Pearson correlation, Cronbach’s alpha values, and Mc Donald’s Omega in the four-factor model. The results showed satisfying item-to-rest Pearson correlations for all items where the lowest correlation values were found for item n5 (0.36) and n6 (0.38), respectively, while the remaining values were above 0.55. The new dimension had a Cronbach’s alpha of 0.88, while Cronbach’s alpha was 0.68 for Action, 0.83 for Diversion, and 0.78 for Resignation. Similar results were found with Mc Donald’s Omega ranging from 0.70 for Action to 0.88 for Verbalisation.

Table 6

Internal consistency in the four-factor model by item-to-rest Pearson correlation and Cronbach’s Alpha (n = 510).

Test–retest agreement and reliability

Figure 4 shows the BA plots including the observed average agreement and the 95% limits of agreement for the test–retest of the BACQ-14. The upper and lower limits of agreement are reported in Table 7 which also shows Pearson correlation coefficients, intraclass correlation coefficients (ICC), concordance correlation coefficients Rho, average difference between measurements and the standard deviation of the difference, standard error of measurement, and the coefficient of repeatability. For all subscales, the BA plots showed homogeneous but substantial variation across the measurement range, thus no systematic differences were observed. The observed average agreement for Verbalisation was −0.20 (SD: 2.68). The 95% limits of agreement ranged from −5.46 to 5.06, which means that the difference between the two measurement points for 95% of the sample lies within this interval.

Figure 4

Table 7

Test–retest agreement and reliability (N = 209).

ICC, intra class correlation (ICC); Diff, the average difference between measurements; SEM, standard error of measurement (agreement); CR, coefficient of repeatability; LOA, limits of agreement (Bland–Altman plot).

The observed average agreement for Action was −0.43 (SD: 2.21) with 95% limits of agreement of [−4.77; 3.90], for Diversion it was −0.30 (SD: 2.68) with 95% limits of agreement of [−5.55; 4.95], and for Resignation it was −0.17 (SD: 2.22) with 95% limits of agreement of [−4.52;4.18].

Regarding reliability, the estimated ICC was 0.64 for Diversion and ranged between 0.75 to 0.77 for Resignation, Verbalisation and Action, yielding very similar estimates as Lin’s concordance coefficient Rho (Table 7).

The BA plots regarding the test–retest of BACQ-12 showed similar results as the BACQ-14 for Approach, Diversion, and Resignation respectively, see Supplementary Figure 1.

Hypotheses testing

Table 8 shows the mean sum scores of the BACQ-14 when using the scale as a four-factor model. The results for Verbalisation showed that women/other had a higher mean sum score (15.48, 95% CI: 15.04;15.92)) than men (13.85, 95% CI: 13.36;14.35), and that respondents with high educational level had a higher mean sum score (15.39, 95% CI: 14.92;15.86) than respondents with no/other education (11.46, 95% CI: 9.46;13.46). Additionally, all three pre-specified hypotheses regarding Verbalisation were confirmed (p-values <0.05), e.g., higher Verbalisation scores were reported amongst individuals with excellent self-rated health (15.23, 95% CI: 13.96;16.51) and very good physical fitness (15.29, 95% CI: 14.21;16.38) compared to individuals with poor self-rated health (12.95, 95% CI: 10.90;15.00) and poor physical fitness (12.96, 95% CI: 11.64;14.28). Additionally, lower scores on Verbalisation were reported amongst individuals being extremely concerned about current health (13.56, 95% CI: 9.58;17.53) compared to individuals being not at all concerned for their health (15.12, 95% CI: 14.46;15.77). For comparison across subscales, means and corresponding 95% CI for both the raw scores, the transformed scores, and the z-scores are found in Supplementary Table 4.

Table 8

Mean (standard deviation) and corresponding 95% confidence intervals (CI) for the sum scores of each coping dimension in the four-factor model (n = 510).

p-values are based on ANOVA (linear regression) models, separately for each score and factor, and they are listed in italics in the score columns. Additionally, z-score means (based on the standard deviation in the whole sample) are presented by factor.

CI, confidence interval. Statistical significance p-value <0.05. ^*Less than 5 reported “Other” for sex. Therefore, these respondents are grouped with women, which is the largest group.

Discussion

The BACQ-12 was successfully revised and expanded to the BACQ-14, which included a new dimension called Verbalisation. In this process, two previous items were replaced with four new items covering both emotion-based and problem-based coping strategies. Furthermore, the introductory text was revised. The pilot tests showed good content validity with good comprehensibility of all items. The item distribution showed some ceiling effects of the items in Verbalisation with response rates for completely agree ranging from 18 to 36%. The EFA showed that both a three-factor model and a four-factor model could be appropriate, the four-factor structure was substantially better supported than the three-factor structure and met most, but not all, prespecified fit criteria. Cronbach’s alpha for Verbalisation was 0.88, indicating high internal consistency in the new domain. The BACQ-14 was able to detect differences between sex and subgroups concerning health-related factors. For example, Verbalisation scores were statistically higher amongst women/other and individuals with lower concern for their health, higher self-rated health, and better self-reported physical fitness as hypothesised. Lastly, the BACQ-14 had some challenges with measurement error, as the 95% limits of agreements in the BA plots indicated large random variation.

Discussion of results and comparison with other studies

The results show that the BACQ-14 can be used as both a three-factor and a four-factor model, but the four-factor model had a substantially better fit and met most of the prespecified fit criteria, yet not all. In the BACQ-14, Diversion and Resignation were considered as two separate factors as proposed in the validation of the BACQ-12 (Sætre et al., 2025). N12 and n13 loaded on both factor III and IV, but although it is meaningful to combine Diversion and Resignation in a single factor because they both reflect Avoidance, it may be useful to distinguish between Diversion and Resignation when targeting coping, e.g., in psychotherapy (Frølund Pedersen et al., 2016; Schröder et al., 2012). Retaining Diversion and Resignation as two separate factors will also allow comparison with previous studies measuring Diversion and Resignation separately (Loton et al., 2016; Raasthøj et al., 2022; Polman et al., 2010).

In the four-factor model, Verbalisation had higher item-to-rest correlation and Cronbach’s alpha than the other dimensions of the BACQ-14, indicating good internal consistency. The poorest performing dimension was Action. When scrutinising the wording of these items, they seem more heterogenic compared to the rest of the scale as they cover both problem-based coping, positive thinking, and the individual’s attitude towards physical exercise which can both reflect coping (e.g., some people may go for a run when experiencing stress) and the individual’s overall health behaviour. All these aspects are difficult to cover while keeping briefness of the scale. Importantly, some cognitive efforts such as positive thinking may be unconscious, and hence the term Action also includes such strategies. Although the internal consistency is lower than for the remaining domains, it is considered acceptable and similar to other scales such as the brief COPE (Carver, 1997).

Generally, psychologists debate whether coping should be understood primarily as a trait, that is, a relatively stable characteristics related to personality, or as a state, meaning that it varies according to the situation (Folkman et al., 1986; Connor-Smith and Flachsbart, 2007; Greenaway et al., 2015). Coping is a very complex concept entangled by sensations, thoughts, emotions, and overall behaviour as suggested in the Common-Sense Model by Leventhal et al. (2016), which makes it inherently challenging to measure. In our test–retest analysis, we could not establish whether true within-person change had occurred during the two- to three-week interval between the first and second administration. Consequently, participant stability must be regarded as an assumption underlying the analysis. The test–retest assessment of the BACQ-12 and BACQ-14 demonstrated an absolute agreement over a two-week interval, with mean differences close to zero across all four factors. However, the 95% limits of agreement were approximately ±5 points, suggesting that changes within this range may be attributable to random measurement error, assuming participant stability. However, the retest surveys did not include a stability indicator, which limits the interpretation of these findings, as some of the changes in coping scores may reflect true changes rather than measurement error. This degree of measurement error has important implications. Clinically, it may limit the instrument’s sensitivity to detect small yet potentially relevant changes in coping at the individual level, thereby constraining its utility for monitoring patient progress. From a research perspective, measurement error of this magnitude may attenuate observed associations and reduce statistical power, increasing the risk of underestimating true effects. Whether this level of variability is inherent to coping constructs or reflects limitations of the instrument remains uncertain. To our knowledge, no prior test–retest evaluation of the BACQ-12 has been conducted, and existing test–retest studies of the widely used Brief COPE have been limited to specific populations, such as women with breast cancer, which are not directly comparable to the general population (Yusoff et al., 2010; Rodrigues et al., 2022).

Strengths and limitations

Strengths of the study include that development of the BACQ-14 and the evaluation of the clinimetric properties followed the COSMIN guidelines (Mokkink et al., 2010; de Vet, 2011). The constructs of Verbalisation were defined in an expert panel, and a thorough developmental process was carried out to ensure that different aspects of coping were included. The proposed items showed good content validity in the pilot tests.

Participants were recruited through a research agency, and we were therefore unable to obtain a fully representative sample of the Danish general population. However, a substantial effort was done to ensure diversity in the study population. The three subsamples used for analyses showed equal distribution of sex and several other sociodemographics, which increases the generalisability. The use of a research agency for the field test also led to the inclusion of both healthy and non-healthy individuals in the study sample, which is a strength as living with a chronic disease may affect how you cope and which situations you think of when completing a questionnaire (Suls and Fletcher, 1985). However, more individuals aged 60–79 years and less individuals aged 20–39 years comprised the Retest-sample compared to sample A and B. Overall, we cannot rule out the risk of selection bias, neither for the initial recruitment or for the retest procedure. Individuals who are included in the database of a research agency may have a special interest in health behaviours which can be related to their coping profile, and individuals who responded twice may be slightly more approach and persistent in their coping behaviour compared to others. Despite the efforts done to reduce selection bias, these bias limits the external validity and the generalizability to the general population.

Limitations include that the study is carried out in a Danish setting. Cultural context is important for coping processes (Chun et al., 2006), and although the results may be applicable in settings resembling the Danish culture, the external validity may be limited. Proper translation and cultural adaption must be done, and the cross-cultural validity should be examined before the instrument can be used in other countries (Mokkink et al., 2010). Also, the briefness of the scale and the reference to general coping rather than situational coping, limit the possibility to give profound knowledge about coping in specific situations. For these purposes, other tools may be more applicable. It should also be noted that due to the small number of respondents reporting other for sex, the results for these individuals are inconclusive as they are merged with the results for women comprising the majority of the group. Finally, development and test of the scale have so far been limited to research purposes. Whether and how the BACQ-14 can be used in clinical settings needs to be explored further.

Implications

The BACQ-14 provides a brief measure of general coping reflecting two aspects of both approach and avoidant coping. This makes the BACQ-14 suitable for inclusion in comprehensive questionnaire studies where a brief measure of general coping is warranted amongst several constructs. The expansion of the BACQ-12 to the BACQ-14 allows a more nuanced evaluation of approach which may be of importance in both academic and clinical settings. However, it is important to note that the BACQ-14 demonstrated large random variation which limits the instrument’s sensitivity to detect small yet potentially clinically relevant changes in coping at the individual level. Consequently, the BACQ-14 has limited abilities for monitoring patient progress.

Conclusion

After a thorough developmental process, the BACQ-14 was successfully established based on the BACQ-12 where two items were replaced with four new items in a dimension called Verbalisation. The validation procedure demonstrated that the BACQ-14 had good content validity and acceptable internal consistency, although some variation was found between the four dimensions. The new dimension Verbalisation had the highest Cronbach Alpha, whereas the other approach dimension Action showed the lowest internal consistency. Based on the factor analyses and clinical relevance, we recommend using the BACQ-14 as a four-factor model including the four dimensions: Verbalisation, Action, Diversion, and Resignation. With regard to test–retest-reliability, the BACQ-14 showed large random variation which should be considered if aiming to monitor coping over time.

Statements

Data availability statement

The datasets generated and analysed in the current study are not publicly available and cannot be shared due to the data protection regulations of the Danish Data Protection Agency. Access to data is strictly limited to the researchers who have obtained permission for data processing. The permission was given to the Research Unit of General Practice, Department of Public Health, University of Southern Denmark, Odense, Denmark. Further inquiries can be made to PI Dorte Jarbøl (email: DJarbol@health.sdu.dk). Requests to access the datasets should be directed to Dorte Jarbøl, DJarbol@health.sdu.dk.

Ethics statement

The studies involving humans were approved by The Research Ethics Committee at University of Southern Denmark (Case no. 21/29156) and registered by the Danish Data Protection Agency (j.no. 2011-41-6651) through the Research and Innovation Organisation (RIO), University of Southern Denmark (Project number 10.104). The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study.

Author contributions

IH: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Writing – original draft, Writing – review & editing. DJ: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Supervision, Validation, Writing – original draft, Writing – review & editing. SW: Conceptualization, Data curation, Formal analysis, Methodology, Software, Writing – original draft, Writing – review & editing. TC: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Supervision, Writing – original draft, Writing – review & editing. KB: Conceptualization, Formal analysis, Investigation, Methodology, Writing – original draft, Writing – review & editing. LS: Conceptualization, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Writing – original draft, Writing – review & editing.

Funding

The author(s) declared that financial support was received for this work and/or its publication. This work was supported by the Danish Cancer Society under Grant R321-A17518 and Region of Southern Denmark under Grant A891. The funding sources did not have any influence on the design of the study.

Acknowledgments

We would like to thank all respondents of both the initial and follow-up questionnaire. Also, thanks to Helle Obenhausen Andersen and Dorrie Lee Jones who translated the BACQ-14, and another thanks to Helle Obenhausen Andersen who has also proofread the final manuscript, and to Emma Nielsen who assisted with the data collection.

Conflict of interest

The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The author(s) declared that Generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fpsyg.2026.1766709/full#supplementary-material

References

• 1

  AllenM. T. (2021). Explorations of avoidance and approach coping and perceived stress with a computer-based avatar task: detrimental effects of resignation and withdrawal. Peerj9:e11265. doi: 10.7717/peerj.11265,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 2

  AltmanD. G.BlandJ. M. (1983). Measurement in medicine: the analysis of method comparison studies. J. Roy. Statist. Soc. Ser. D32, 307–317. doi: 10.2307/2987937

  • CrossRef
  • Google Scholar

• 3

  AmirkhanJ. (1990). A factor analytically derived measure of coping: the coping strategy indicator. J. Pers. Soc. Psychol.59, 1066–1074. doi: 10.1037/0022-3514.59.5.1066

  • CrossRef
  • Google Scholar

• 4

  BeatonD. E.BombardierC.GuilleminF.FerrazM. B. (1976). Guidelines for the process of cross-cultural adaptation of self-report measures. Spine (Phila Pa 1976)25, 3186–3191. doi: 10.1097/00007632-200012150-00014,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 5

  BüssingA.WalachH.KohlsN.ZimmermannF.TrousselardM. (2013). Conscious presence and self control as a measure of situational awareness in soldiers - a validation study. Int J Ment Health Syst7:1. doi: 10.1186/1752-4458-7-1,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 6

  CarverC. S. (1997). You want to measure coping but your protocol’ too long: consider the brief cope. Int. J. Behav. Med.4, 92–100. doi: 10.1207/s15327558ijbm0401_6,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 7

  CarverC. S.Connor-SmithJ. (2010). Personality and coping. Annu. Rev. Psychol.61, 679–704. doi: 10.1146/annurev.psych.093008.100352

  • CrossRef
  • Google Scholar

• 8

  CarverC. S.ScheierM. F.WeintraubJ. K. (1989). Assessing coping strategies: a theoretically based approach. J. Pers. Soc. Psychol.56, 267–283. doi: 10.1037//0022-3514.56.2.267,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 9

  ChunC.-A.MoosR. H.et al. (2006). Culture: A Fundamental Context for the Stress and Coping Paradigm. Boston: Springer US, 29–53.

  • Google Scholar

• 10

  Connor-SmithJ. K.FlachsbartC. (2007). Relations between personality and coping: a meta-analysis. J. Pers. Soc. Psychol.93, 1080–1107. doi: 10.1037/0022-3514.93.6.1080,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 11

  de VetH. C. W. (2011). Measurement in Medicine: A Practical Guide. 1st Edn Cambridge: Cambridge University Press.

  • Google Scholar

• 12

  DinnoA.Gently Clarifying the Application of Horn's Parallel Analysis to Principal Component Analysis Versus Factor Analysis. (2010)

  • Google Scholar

• 13

  EndlerN.ParkerJ. (1994). Assessment of multidimensional coping: task, emotion, and avoidance strategies. Psychol. Assess.6, 50–60. doi: 10.1037/1040-3590.6.1.50

  • CrossRef
  • Google Scholar

• 14

  FinsetA.SteineS.HaugliL.SteenE.LaerumE. (2002). The brief approach/avoidance coping questionnaire: development and validation. Psychol. Health Med.7, 75–85. doi: 10.1080/13548500120101577

  • CrossRef
  • Google Scholar

• 15

  FolkmanS.LazarusR. S. (1988). Manual for the ways of Coping Questionnaire. Research ed Edn Palo Alto: Consulting Psychologists Press.

  • Google Scholar

• 16

  FolkmanS.LazarusR. S.GruenR. J.DeLongisA. (1986). Appraisal, coping, health status, and psychological symptoms. J. Pers. Soc. Psychol.50, 571–579. doi: 10.1037/0022-3514.50.3.571,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 17

  Frølund PedersenH.FrostholmL.Søndergaard JensenJ.ØrnbølE.SchröderA. (2016). Neuroticism and maladaptive coping in patients with functional somatic syndromes. Br. J. Health Psychol.21, 917–936. doi: 10.1111/bjhp.12206,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 18

  GreenawayK. H.LouisW. R.et al. (2015). “Chapter 12 - measures of coping for psychological well-being,” in Measures of Personality and Social Psychological Constructs, eds. BoyleG. J.SaklofskeD. H.MatthewsG. (San Diego: Academic Press), 322–351.

  • Google Scholar

• 19

  HairJ. F.AndersonR. E.TathamR. L.BlackW. (1998). Multivariate data analysis, vol. 5th. 5th Edn London: Prentice-Hall.

  • Google Scholar

• 20

  HooperD.CoughlanJ.MullenM. (2007). Structural equation modeling: guidelines for determining model fit. Electron. J. Bus. Res. Methods6, 53–60.

  • Google Scholar

• 21

  JacobsJ. M.CarverC. S. (2020). “Personality and coping,” in The Wiley Encyclopedia of Health Psychology, eds. SweenyK.RobbinsM. L.CohenL. M. (Chichester: John Wiley & Sons Ltd.), 411–420.

  • Google Scholar

• 22

  KaiserH. F.RiceJ. (1974). Little jiffy, mark Iv. Educ. Psychol. Meas.34, 111–117. doi: 10.1177/001316447403400115

  • CrossRef
  • Google Scholar

• 23

  KirmayerL. J. (2005). Culture, context and experience in psychiatric diagnosis. Psychopathology38, 192–196. doi: 10.1159/000086090,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 24

  KirmayerL. J.SartoriusN. (2007). Cultural models and somatic syndromes. Psychosom. Med.69, 832–840. doi: 10.1097/PSY.0b013e31815b002c,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 25

  KlineR. B. (2005). Principles and Practice of Structural Equation Modeling. 2nd Edn. New York: Guildford Press.

  • Google Scholar

• 26

  KrohneH. W.EgloffB.VarnerL. J.BurnsL. R.WeidnerG.EllisH. C. (2000). The assessment of dispositional vigilance and cognitive avoidance: factorial structure, psychometric properties, and validity of the Mainz coping inventory. Cogn. Ther. Res.24, 297–311. doi: 10.1023/A:1005511320194

  • CrossRef
  • Google Scholar

• 27

  LazarusR.FolkmanS. (1984). Stress, Appraisal and Coping. New York: Springer Publishing Company Inc., 117–178.

  • Google Scholar

• 28

  LedesmaR. D.Valero-MoraP.MacbethG. (2015). The scree test and the number of factors: a dynamic graphics approach. Span. J. Psychol.18:E11. doi: 10.1017/sjp.2015.13,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 29

  LeventhalH.PhillipsL. A.BurnsE. (2016). The common-sense model of self-regulation (CSM): a dynamic framework for understanding illness self-management. J. Behav. Med.39, 935–946. doi: 10.1007/s10865-016-9782-2,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 30

  LinL. I. K. (1989). A concordance correlation coefficient to evaluate reproducibility. Biometrics45, 255–268. doi: 10.2307/2532051,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 31

  LotonD.BorkolesE.LubmanD.PolmanR. (2016). Video game addiction, engagement and symptoms of stress, depression and anxiety: the mediating role of coping. Int. J. Ment. Health Addict.14, 565–578. doi: 10.1007/s11469-015-9578-6

  • CrossRef
  • Google Scholar

• 32

  MokkinkL. B.TerweeC. B.PatrickD. L.AlonsoJ.StratfordP. W.KnolD. L.et al. (2010b). The COSMIN checklist for assessing the methodological quality of studies on measurement properties of health status measurement instruments: an international Delphi study. Qual. Life Res.19, 539–549. doi: 10.1007/s11136-010-9606-8,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 33

  MokkinkL. B.TerweeC. B.PatrickD. L.AlonsoJ.StratfordP. W.KnolD. L.et al. (2010). The COSMIN study reached international consensus on taxonomy, terminology, and definitions of measurement properties for health-related patient-reported outcomes. J. Clin. Epidemiol.63, 737–745. doi: 10.1016/j.jclinepi.2010.02.006,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 34

  MoskowitzJ. T.HultJ. R.BussolariC.AcreeM. (2009). What works in coping with HIV? A meta-analysis with implications for coping with serious illness. Psychol. Bull.135, 121–141. doi: 10.1037/a0014210,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 35

  PanayiotouG.KareklaM.MeteI. (2014). Dispositional coping in individuals with anxiety disorder symptomatology: avoidance predicts distress. J. Context. Behav. Sci.3, 314–321. doi: 10.1016/j.jcbs.2014.07.001

  • CrossRef
  • Google Scholar

• 36

  PenleyJ. A.TomakaJ.WiebeJ. S. (2002). The association of coping to physical and psychological health outcomes: a meta-analytic review. J. Behav. Med.25, 551–603. doi: 10.1023/a:1020641400589,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 37

  PolmanR.BorkolesE.NichollsA. R. (2010). Type D personality, stress, and symptoms of burnout: the influence of avoidance coping and social support. Br. J. Health Psychol.15, 681–696. doi: 10.1348/135910709x479069,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 38

  QinS.NelsonL.McLeodL.EremencoS.CoonsS. J. (2019). Assessing test–retest reliability of patient-reported outcome measures using intraclass correlation coefficients: recommendations for selecting and documenting the analytical formula. Qual. Life Res.28, 1029–1033. doi: 10.1007/s11136-018-2076-0,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 39

  RaasthøjI.RasmussenS.CarstensenT. B. W.WehbergS.RosendalM.JarbølD. E. (2022). Coping strategies among individuals with multiple physical symptoms: a general population-based cross-sectional study. J. Psychosom. Res.165:111140. doi: 10.1016/j.jpsychores.2022.111140

  • CrossRef
  • Google Scholar

• 40

  RasmussenS.SondergaardJ.LarsenP. V.BalasubramaniamK.ElnegaardS.SvendsenR. P.et al. (2014). The Danish symptom cohort: questionnaire and feasibility in the nationwide study on symptom experience and healthcare-seeking among 100 000 individuals. Int. J. Fam. Med.2014:187280. doi: 10.1155/2014/187280,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 41

  RodriguesF.FigueiredoN.RodriguesJ.FerreiraR.Hernández-MendoA.MonteiroD. (2022). A comprehensive review and Bifactor modeling analysis of the brief COPE. Inquiry59:469580221108127. doi: 10.1177/00469580221108127,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 42

  RutskijR.GaardenT.BremnesR.DahlO.FinsetA.FossaS. D.et al. (2010). A study of coping in long-term testicular cancer survivors. Psychol. Health Med.15, 146–158. doi: 10.1080/13548501003623955

  • CrossRef
  • Google Scholar

• 43

  RutterC. L.RutterD. R. (2002). Illness representation, coping and outcome in irritable bowel syndrome (IBS). Br. J. Health Psychol.7, 377–391. doi: 10.1348/135910702320645372,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 44

  SætreL. M. S.RaasthøjI. P.JarbølD. E.BalasubramaniamK.WehbergS.CarstensenT. B.et al. (2025). Coping in the Danish general population: psychometric properties of the Danish version of the brief approach/avoidance coping questionnaire. Psychol. Health, 40, 254–271. doi: 10.1080/08870446.2023.2215263

  • CrossRef
  • Google Scholar

• 45

  SætreL. M. S.RaasthøjI.LauridsenG. B.BalasubramaniamK.HaastrupP.RasmussenS.et al. (2024). Revisiting the symptom iceberg based on the Danish symptom cohort – symptom experiences and healthcare-seeking behaviour in the general Danish population in 2022. Heliyon10:e31090. doi: 10.1016/j.heliyon.2024.e31090,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 46

  SætreL. M. S.RaasthøjI. P.et al. (2023). Revisiting the symptom iceberg based on the Danish symptom cohort – symptom experiences and healthcare-seeking behavior in the general Danish population in 2022. Heliyon.

  • Google Scholar

• 47

  SchröderA.RehfeldE.ØrnbølE.SharpeM.LichtR. W.FinkP. (2012). Cognitive–behavioural group treatment for a range of functional somatic syndromes: randomised trial. Br. J. Psychiatry200, 499–507. doi: 10.1192/bjp.bp.111.098681,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 48

  SkinnerE.EdgeK.SkinnerE. A.AltmanJ.SherwoodH. (2003). Searching for the structure of coping: a review and critique of category systems for classifying ways of coping. Psychol. Bull.129, 216–269. doi: 10.1037/0033-2909.129.2.216

  • CrossRef
  • Google Scholar

• 49

  SobergH. L.Bautz-HolterE.FinsetA.RoiseO.AndelicN. (2015). Physical and mental health 10 years after multiple trauma: a prospective cohort study. J. Trauma Acute Care Surg.78, 628–633. doi: 10.1097/ta.0000000000000541,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 50

  SobergH. L.FinsetA.Bautz-HolterE.SandvikL.RoiseO. (2007). Return to work after severe multiple injuries: a multidimensional approach on status 1 and 2 years postinjury. J. Trauma62, 471–481. doi: 10.1097/TA.0b013e31802e95f4

  • CrossRef
  • Google Scholar

• 51

  StataCorp (2019). Stata Statistical Software. College Station: StataCorp LLC.

  • Google Scholar

• 52

  SulsJ.FletcherB. (1985). The relative efficacy of avoidant and nonavoidant coping strategies: a meta-analysis. Health Psychol.4, 249–288. doi: 10.1037//0278-6133.4.3.249,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 53

  Syddansk Universitet. Wording of the BACQ-12, the BACQ-14, and additional questions of the questionnaire. (2024). Available online at: https://findresearcher.sdu.dk/ws/portalfiles/portal/259027497/BACQ-14_Questionnaire.pdf (Accessed September 1, 2026).

  • Google Scholar

• 54

  TaylorS. E.StantonA. L. (2007). Coping resources, coping processes, and mental health. Annu. Rev. Clin. Psychol.3, 377–401. doi: 10.1146/annurev.clinpsy.3.022806.091520,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 55

  TerweeC. B.KnolD. L.et al. (2011). Development of a Measurement Instrument, in Measurement in Medicine: A Practical Guide. Cambridge: Cambridge University Press, 30–64.

  • Google Scholar

• 56

  UlrichsenM.KristoffersenA. E.KjekenI.HaugenI. K. (2021). Characteristics of persons with hand osteoarthritis visiting complementary and alternative medicine providers. Osteoarthr. Cartil. Open.3:100220. doi: 10.1016/j.ocarto.2021.100220,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 57

  WhitmanR. H. (1929). A short scale for measuring introversion-extroversion. J. Appl. Psychol.13, 499–504. doi: 10.1037/h0070852

  • CrossRef
  • Google Scholar

• 58

  YusoffN.LowW. Y.YipC. H. (2010). Reliability and validity of the brief COPE scale (English version) among women with breast cancer undergoing treatment of adjuvant chemotherapy: a Malaysian study. Med J Malaysia65, 41–44.

  • Google Scholar

Keywords

12-item brief approach/avoidance coping questionnaire, BACQ, clinimetrics, coping, test–retest, validation

Citation

Holst IPR, Jarbøl DE, Wehberg S, Carstensen TBW, Balasubramaniam K and Sætre LMS (2026) The clinimetric properties of an extended version of the brief approach/avoidance coping questionnaire in the Danish population: the BACQ-14. Front. Psychol. 17:1766709. doi: 10.3389/fpsyg.2026.1766709

Received

12 December 2025

Revised

25 September 2026

Accepted

05 October 2026

Published

15 October 2026

Volume

17 - 2026

Updates

Copyright

© 2026 Holst, Jarbøl, Wehberg, Carstensen, Balasubramaniam and Sætre.

This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

*Correspondence: Isabella P. R. Holst, iraasthoj@health.sdu.dk

Disclaimer

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.